[mpich-discuss] *** glibc detected *** mpiexec: munmap_chunk(): invalid pointer: 0x00007fff1d578184 ***
Pavan Balaji
balaji at mcs.anl.gov
Sat Nov 6 09:45:25 CDT 2010
Yes, a bunch of memory leaks got fixed after 1.3. We are working on
releasing 1.3.1 soon (before SC), which should have all these fixes.
-- Pavan
On 11/06/2010 09:31 AM, Jeff Hammond wrote:
> MPICH2 trunk updated yesterday and compiled with GCC 4.5 works fine
> for both MPI tests that failed before so it seems that the error was
> fixed after the 1.3 release.
>
> Jeff
>
> On Sat, Nov 6, 2010 at 9:11 AM, Jeff Hammond<jhammond at mcs.anl.gov> wrote:
>> Sorry for not including this last time, but valgrind seems to indicate
>> the problem is in Hydra.
>>
>> Jeff
>>
>> ==1525== Invalid free() / delete / delete[]
>> ==1525== at 0x4C243AF: free (in
>> /usr/lib64/valgrind/amd64-linux/vgpreload_memcheck.so)
>> ==1525== by 0x421983: HYDT_bsci_finalize (bsci_finalize.c:21)
>> ==1525== by 0x40A07E: HYD_pmci_finalize (pmiserv_pmci.c:371)
>> ==1525== by 0x40537E: main (mpiexec.c:327)
>> ==1525== Address 0x7ff000149 is on thread 1's stack
>> ==1525==
>> ==1525== ERROR SUMMARY: 1 errors from 1 contexts (suppressed: 2 from 1)
>> ==1525== malloc/free: in use at exit: 16,396 bytes in 3 blocks.
>> ==1525== malloc/free: 1,698 allocs, 1,696 frees, 118,394 bytes allocated.
>> ==1525== For counts of detected errors, rerun with: -v
>> ==1525== searching for pointers to 3 not-freed blocks.
>> ==1525== checked 184,080 bytes.
>> ==1525==
>> ==1525==
>> ==1525== 6 bytes in 1 blocks are still reachable in loss record 1 of 3
>> ==1525== at 0x4C256AE: malloc (in
>> /usr/lib64/valgrind/amd64-linux/vgpreload_memcheck.so)
>> ==1525== by 0x5EB1261: strdup (in /lib64/libc-2.9.so)
>> ==1525== by 0x41EA50: HYDT_bind_init (bind.c:44)
>> ==1525== by 0x4096BE: HYD_pmci_launch_procs (pmiserv_pmci.c:293)
>> ==1525== by 0x4050B0: main (mpiexec.c:289)
>> ==1525==
>> ==1525==
>> ==1525== 6 bytes in 1 blocks are definitely lost in loss record 2 of 3
>> ==1525== at 0x4C256AE: malloc (in
>> /usr/lib64/valgrind/amd64-linux/vgpreload_memcheck.so)
>> ==1525== by 0x5EB1261: strdup (in /lib64/libc-2.9.so)
>> ==1525== by 0x41EFFB: HYDT_bind_init (bind.c:40)
>> ==1525==
>> ==1525==
>> ==1525== 16,384 bytes in 1 blocks are still reachable in loss record 3 of 3
>> ==1525== at 0x4C256AE: malloc (in
>> /usr/lib64/valgrind/amd64-linux/vgpreload_memcheck.so)
>> ==1525== by 0x41EB08: HYDT_bind_init (bind.c:53)
>> ==1525== by 0x4096BE: HYD_pmci_launch_procs (pmiserv_pmci.c:293)
>> ==1525== by 0x4050B0: main (mpiexec.c:289)
>> ==1525==
>> ==1525== LEAK SUMMARY:
>> ==1525== definitely lost: 6 bytes in 1 blocks.
>> ==1525== possibly lost: 0 bytes in 0 blocks.
>> ==1525== still reachable: 16,390 bytes in 2 blocks.
>> ==1525== suppressed: 0 bytes in 0 blocks.
>>
>>
>>
>> On Sat, Nov 6, 2010 at 9:08 AM, Jeff Hammond<jhammond at mcs.anl.gov> wrote:
>>> Yep, still happening when I use MPICH2-1.3 pulled via SVN yesterday
>>> and compiled with GCC 4.6. The test that fails is extremely simple.
>>>
>>> Jeff
>>>
>>>
>>> jeff at localhost:~/eclipse/a1/trunk/benchmarks/mpi> cat MPI_simple.c
>>> #include<stdio.h>
>>> #include<mpi.h>
>>>
>>> int main(int argc, char **argv)
>>> {
>>> int provided;
>>> int rank, size;
>>> MPI_Init_thread(&argc,&argv, MPI_THREAD_SINGLE,&provided);
>>> MPI_Comm_rank(MPI_COMM_WORLD,&rank);
>>> MPI_Comm_size(MPI_COMM_WORLD,&size);
>>> MPI_Barrier(MPI_COMM_WORLD);
>>> printf("before MPI_Finalize\n"); fflush(stdout);
>>> MPI_Finalize();
>>> printf("after MPI_Finalize\n"); fflush(stdout);
>>> return(0);
>>> }
>>> jeff at localhost:~/eclipse/a1/trunk/benchmarks/mpi>
>>> /software/mpich/mpich2-1.3-gnu46-build/bin/mpiexec -n 2 MPI_simple.x
>>> before MPI_Finalize
>>> after MPI_Finalize
>>> before MPI_Finalize
>>> after MPI_Finalize
>>> *** glibc detected ***
>>> /software/mpich/mpich2-1.3-gnu46-build/bin/mpiexec: munmap_chunk():
>>> invalid pointer: 0x00007fffd654016b ***
>>> ======= Backtrace: =========
>>> /lib64/libc.so.6[0x7f1cca7040c8]
>>> /software/mpich/mpich2-1.3-gnu46-build/bin/mpiexec[0x421984]
>>> /software/mpich/mpich2-1.3-gnu46-build/bin/mpiexec[0x40a07f]
>>> /software/mpich/mpich2-1.3-gnu46-build/bin/mpiexec[0x40537f]
>>> /lib64/libc.so.6(__libc_start_main+0xe6)[0x7f1cca6ae586]
>>> /software/mpich/mpich2-1.3-gnu46-build/bin/mpiexec[0x404359]
>>> ======= Memory map: ========
>>> 00400000-00449000 r-xp 00000000 08:03 312441
>>> /software/mpich/mpich2-1.3-gnu46-build/bin/mpiexec.hydra
>>> 00648000-00649000 r--p 00048000 08:03 312441
>>> /software/mpich/mpich2-1.3-gnu46-build/bin/mpiexec.hydra
>>> 00649000-0064b000 rw-p 00049000 08:03 312441
>>> /software/mpich/mpich2-1.3-gnu46-build/bin/mpiexec.hydra
>>> 0064b000-00672000 rw-p 0064b000 00:00 0 [heap]
>>> 7f1cca479000-7f1cca48e000 r-xp 00000000 08:02 1417264
>>> /lib64/libgcc_s.so.1
>>> 7f1cca48e000-7f1cca68e000 ---p 00015000 08:02 1417264
>>> /lib64/libgcc_s.so.1
>>> 7f1cca68e000-7f1cca68f000 r--p 00015000 08:02 1417264
>>> /lib64/libgcc_s.so.1
>>> 7f1cca68f000-7f1cca690000 rw-p 00016000 08:02 1417264
>>> /lib64/libgcc_s.so.1
>>> 7f1cca690000-7f1cca7df000 r-xp 00000000 08:02 1417222
>>> /lib64/libc-2.9.so
>>> 7f1cca7df000-7f1cca9df000 ---p 0014f000 08:02 1417222
>>> /lib64/libc-2.9.so
>>> 7f1cca9df000-7f1cca9e3000 r--p 0014f000 08:02 1417222
>>> /lib64/libc-2.9.so
>>> 7f1cca9e3000-7f1cca9e4000 rw-p 00153000 08:02 1417222
>>> /lib64/libc-2.9.so
>>> 7f1cca9e4000-7f1cca9e9000 rw-p 7f1cca9e4000 00:00 0
>>> 7f1cca9e9000-7f1cca9ff000 r-xp 00000000 08:02 1417248
>>> /lib64/libpthread-2.9.so
>>> 7f1cca9ff000-7f1ccabff000 ---p 00016000 08:02 1417248
>>> /lib64/libpthread-2.9.so
>>> 7f1ccabff000-7f1ccac00000 r--p 00016000 08:02 1417248
>>> /lib64/libpthread-2.9.so
>>> 7f1ccac00000-7f1ccac01000 rw-p 00017000 08:02 1417248
>>> /lib64/libpthread-2.9.so
>>> 7f1ccac01000-7f1ccac05000 rw-p 7f1ccac01000 00:00 0
>>> 7f1ccac05000-7f1ccac0d000 r-xp 00000000 08:02 1417496
>>> /lib64/librt-2.9.so
>>> 7f1ccac0d000-7f1ccae0c000 ---p 00008000 08:02 1417496
>>> /lib64/librt-2.9.so
>>> 7f1ccae0c000-7f1ccae0d000 r--p 00007000 08:02 1417496
>>> /lib64/librt-2.9.so
>>> 7f1ccae0d000-7f1ccae0e000 rw-p 00008000 08:02 1417496
>>> /lib64/librt-2.9.so
>>> 7f1ccae0e000-7f1ccae23000 r-xp 00000000 08:02 1417457
>>> /lib64/libnsl-2.9.so
>>> 7f1ccae23000-7f1ccb022000 ---p 00015000 08:02 1417457
>>> /lib64/libnsl-2.9.so
>>> 7f1ccb022000-7f1ccb023000 r--p 00014000 08:02 1417457
>>> /lib64/libnsl-2.9.so
>>> 7f1ccb023000-7f1ccb024000 rw-p 00015000 08:02 1417457
>>> /lib64/libnsl-2.9.so
>>> 7f1ccb024000-7f1ccb026000 rw-p 7f1ccb024000 00:00 0
>>> 7f1ccb026000-7f1ccb07b000 r-xp 00000000 08:02 1417485
>>> /lib64/libm-2.9.so
>>> 7f1ccb07b000-7f1ccb27a000 ---p 00055000 08:02 1417485
>>> /lib64/libm-2.9.so
>>> 7f1ccb27a000-7f1ccb27b000 r--p 00054000 08:02 1417485
>>> /lib64/libm-2.9.so
>>> 7f1ccb27b000-7f1ccb27c000 rw-p 00055000 08:02 1417485
>>> /lib64/libm-2.9.so
>>> 7f1ccb27c000-7f1ccb291000 r-xp 00000000 08:02 1417270
>>> /lib64/libz.so.1.2.3
>>> 7f1ccb291000-7f1ccb490000 ---p 00015000 08:02 1417270
>>> /lib64/libz.so.1.2.3
>>> 7f1ccb490000-7f1ccb491000 r--p 00014000 08:02 1417270
>>> /lib64/libz.so.1.2.3
>>> 7f1ccb491000-7f1ccb492000 rw-p 00015000 08:02 1417270
>>> /lib64/libz.so.1.2.3
>>> 7f1ccb492000-7f1ccb494000 r-xp 00000000 08:02 1417482
>>> /lib64/libdl-2.9.so
>>> 7f1ccb494000-7f1ccb694000 ---p 00002000 08:02 1417482
>>> /lib64/libdl-2.9.so
>>> 7f1ccb694000-7f1ccb695000 r--p 00002000 08:02 1417482
>>> /lib64/libdl-2.9.so
>>> 7f1ccb695000-7f1ccb696000 rw-p 00003000 08:02 1417482
>>> /lib64/libdl-2.9.so
>>> 7f1ccb696000-7f1ccb7ea000 r-xp 00000000 08:02 521675
>>> /usr/lib64/libxml2.so.2.7.1
>>> 7f1ccb7ea000-7f1ccb9e9000 ---p 00154000 08:02 521675
>>> /usr/lib64/libxml2.so.2.7.1
>>> 7f1ccb9e9000-7f1ccb9f1000 r--p 00153000 08:02 521675
>>> /usr/lib64/libxml2.so.2.7.1
>>> 7f1ccb9f1000-7f1ccb9f3000 rw-p 0015b000 08:02 521675
>>> /usr/lib64/libxml2.so.2.7.1
>>> 7f1ccb9f3000-7f1ccb9f4000 rw-p 7f1ccb9f3000 00:00 0
>>> 7f1ccb9f4000-7f1ccba12000 r-xp 00000000 08:02 1420584
>>> /lib64/ld-2.9.so
>>> 7f1ccbbd3000-7f1ccbbd8000 rw-p 7f1ccbbd3000 00:00 0
>>> 7f1ccbc0f000-7f1ccbc11000 rw-p 7f1ccbc0f000 00:00 0
>>> 7f1ccbc11000-7f1ccbc12000 r--p 0001d000 08:02 1420584
>>> /lib64/ld-2.9.so
>>> 7f1ccbc12000-7f1ccbc13000 rw-p 0001e000 08:02 1420584
>>> /lib64/ld-2.9.so
>>> 7fffd651e000-7fffd6541000 rw-p 7ffffffdc000 00:00 0 [stack]
>>> 7fffd65ff000-7fffd6600000 r-xp 7fffd65ff000 00:00 0 [vdso]
>>> ffffffffff600000-ffffffffff601000 r-xp 00000000 00:00 0
>>> [vsyscall]
>>> Aborted
>>>
>>>
>>> On Fri, Nov 5, 2010 at 12:15 PM, Pavan Balaji<balaji at mcs.anl.gov> wrote:
>>>>
>>>> Did you try 1.3? The attached program works fine with trunk.
>>>>
>>>> -- Pavan
>>>>
>>>> On 11/05/2010 11:21 AM, Jeff Hammond wrote:
>>>>>
>>>>> I'm seeing an error when parallel jobs run with MPICH2 1.3a1 terminate.
>>>>>
>>>>> I see the problem with NWChem (attached) but also with the trivial code
>>>>> below.
>>>>>
>>>>> Jeff
>>>>>
>>>>>
>>>>> ==========================================================
>>>>> stdout and stderr
>>>>> ==========================================================
>>>>> jeff at localhost:~/eclipse/a1/trunk/benchmarks/mpi> mpiexec -np 2
>>>>> MPI_Allreduce_tests.x
>>>>> Test 0: sum 1 out-of-place
>>>>> 0: in=1 out=1 ans=1 PASSED
>>>>> Test 1: sum 1 in-place
>>>>> 0: in=1 out=1 ans=1 PASSED
>>>>> Test 2: sum 2 out-of-place
>>>>> 0: in=1 out=2 ans=2 PASSED
>>>>> Test 3: sum 2 in-place
>>>>> 0: in=1 out=2 ans=2 PASSED
>>>>> Test 4: prod 1 out-of-place
>>>>> 0: in=1 out=0 ans=0 PASSED
>>>>> Test 5: prod 1 in-place
>>>>> 0: in=1 out=0 ans=0 PASSED
>>>>> Test 6: prod 2 out-of-place
>>>>> 0: in=1 out=1 ans=1 PASSED
>>>>> Test 7: prod 2 in-place
>>>>> 0: in=1 out=1 ans=1 PASSED
>>>>> 1: in=0 out=1 ans=1 PASSED
>>>>> 1: in=0 out=1 ans=1 PASSED
>>>>> 1: in=1 out=2 ans=2 PASSED
>>>>> 1: in=1 out=2 ans=2 PASSED
>>>>> 1: in=0 out=0 ans=0 PASSED
>>>>> 1: in=0 out=0 ans=0 PASSED
>>>>> 1: in=1 out=1 ans=1 PASSED
>>>>> 1: in=1 out=1 ans=1 PASSED
>>>>> *** glibc detected *** mpiexec: munmap_chunk(): invalid pointer:
>>>>> 0x00007fffcb3ce169 ***
>>>>> ======= Backtrace: =========
>>>>> /lib64/libc.so.6[0x7f5415e5b0c8]
>>>>> mpiexec[0x41908b]
>>>>> mpiexec[0x403a97]
>>>>> /lib64/libc.so.6(__libc_start_main+0xe6)[0x7f5415e05586]
>>>>> mpiexec[0x402f29]
>>>>> ======= Memory map: ========
>>>>> 00400000-00426000 r-xp 00000000 08:03 2426704
>>>>> /software/mpich/mpich2-1.3a1-gnu-build/bin/mpiexec
>>>>> 00625000-00626000 r--p 00025000 08:03 2426704
>>>>> /software/mpich/mpich2-1.3a1-gnu-build/bin/mpiexec
>>>>> 00626000-00627000 rw-p 00026000 08:03 2426704
>>>>> /software/mpich/mpich2-1.3a1-gnu-build/bin/mpiexec
>>>>> 00627000-0064a000 rw-p 00627000 00:00 0
>>>>> [heap]
>>>>> 7f5415bd0000-7f5415be5000 r-xp 00000000 08:02 1417264
>>>>> /lib64/libgcc_s.so.1
>>>>> 7f5415be5000-7f5415de5000 ---p 00015000 08:02 1417264
>>>>> /lib64/libgcc_s.so.1
>>>>> 7f5415de5000-7f5415de6000 r--p 00015000 08:02 1417264
>>>>> /lib64/libgcc_s.so.1
>>>>> 7f5415de6000-7f5415de7000 rw-p 00016000 08:02 1417264
>>>>> /lib64/libgcc_s.so.1
>>>>> 7f5415de7000-7f5415f36000 r-xp 00000000 08:02 1417222
>>>>> /lib64/libc-2.9.so
>>>>> 7f5415f36000-7f5416136000 ---p 0014f000 08:02 1417222
>>>>> /lib64/libc-2.9.so
>>>>> 7f5416136000-7f541613a000 r--p 0014f000 08:02 1417222
>>>>> /lib64/libc-2.9.so
>>>>> 7f541613a000-7f541613b000 rw-p 00153000 08:02 1417222
>>>>> /lib64/libc-2.9.so
>>>>> 7f541613b000-7f5416140000 rw-p 7f541613b000 00:00 0
>>>>> 7f5416140000-7f5416155000 r-xp 00000000 08:02 1417457
>>>>> /lib64/libnsl-2.9.so
>>>>> 7f5416155000-7f5416354000 ---p 00015000 08:02 1417457
>>>>> /lib64/libnsl-2.9.so
>>>>> 7f5416354000-7f5416355000 r--p 00014000 08:02 1417457
>>>>> /lib64/libnsl-2.9.so
>>>>> 7f5416355000-7f5416356000 rw-p 00015000 08:02 1417457
>>>>> /lib64/libnsl-2.9.so
>>>>> 7f5416356000-7f5416358000 rw-p 7f5416356000 00:00 0
>>>>> 7f5416358000-7f5416376000 r-xp 00000000 08:02 1420584
>>>>> /lib64/ld-2.9.so
>>>>> 7f541653a000-7f541653c000 rw-p 7f541653a000 00:00 0
>>>>> 7f5416573000-7f5416575000 rw-p 7f5416573000 00:00 0
>>>>> 7f5416575000-7f5416576000 r--p 0001d000 08:02 1420584
>>>>> /lib64/ld-2.9.so
>>>>> 7f5416576000-7f5416577000 rw-p 0001e000 08:02 1420584
>>>>> /lib64/ld-2.9.so
>>>>> 7fffcb3b9000-7fffcb3cf000 rw-p 7ffffffe9000 00:00 0
>>>>> [stack]
>>>>> 7fffcb3ff000-7fffcb400000 r-xp 7fffcb3ff000 00:00 0
>>>>> [vdso]
>>>>> ffffffffff600000-ffffffffff601000 r-xp 00000000 00:00 0
>>>>> [vsyscall]
>>>>> Aborted
>>>>>
>>>>>
>>>>> ==========================================================
>>>>> source
>>>>> ==========================================================
>>>>> /*
>>>>> * The following is a notice of limited availability of the code, and
>>>>> disclaimer
>>>>> * which must be included in the prologue of the code and in all source
>>>>> listings
>>>>> * of the code.
>>>>> *
>>>>> * Copyright (c) 2010 Argonne Leadership Computing Facility, Argonne
>>>>> National
>>>>> * Laboratory
>>>>> *
>>>>> * Permission is hereby granted to use, reproduce, prepare derivative
>>>>> works, and
>>>>> * to redistribute to others.
>>>>> *
>>>>> *
>>>>> * LICENSE
>>>>> *
>>>>> * Redistribution and use in source and binary forms, with or without
>>>>> * modification, are permitted provided that the following conditions are
>>>>> * met:
>>>>> *
>>>>> * - Redistributions of source code must retain the above copyright
>>>>> * notice, this list of conditions and the following disclaimer.
>>>>> *
>>>>> * - Redistributions in binary form must reproduce the above copyright
>>>>> * notice, this list of conditions and the following disclaimer listed
>>>>> * in this license in the documentation and/or other materials
>>>>> * provided with the distribution.
>>>>> *
>>>>> * - Neither the name of the copyright holders nor the names of its
>>>>> * contributors may be used to endorse or promote products derived from
>>>>> * this software without specific prior written permission.
>>>>> *
>>>>> * The copyright holders provide no reassurances that the source code
>>>>> * provided does not infringe any patent, copyright, or any other
>>>>> * intellectual property rights of third parties. The copyright holders
>>>>> * disclaim any liability to any recipient for claims brought against
>>>>> * recipient by any third party for infringement of that parties
>>>>> * intellectual property rights.
>>>>> *
>>>>> * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
>>>>> * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
>>>>> * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
>>>>> * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
>>>>> * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
>>>>> * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
>>>>> * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
>>>>> * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
>>>>> * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
>>>>> * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
>>>>> * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
>>>>> */
>>>>>
>>>>> #include<stdio.h>
>>>>> #include<stdlib.h>
>>>>> #include<mpi.h>
>>>>>
>>>>> int main(int argc, char **argv)
>>>>> {
>>>>> int provided;
>>>>>
>>>>> int rank, size;
>>>>>
>>>>> int count;
>>>>>
>>>>> int in, out, ans;
>>>>>
>>>>> MPI_Init_thread(&argc,&argv, MPI_THREAD_MULTIPLE,&provided);
>>>>>
>>>>> MPI_Comm_rank(MPI_COMM_WORLD,&rank);
>>>>> MPI_Comm_size(MPI_COMM_WORLD,&size);
>>>>>
>>>>> MPI_Barrier(MPI_COMM_WORLD);
>>>>>
>>>>> /* TESTING SUM OPERATOR */
>>>>>
>>>>> if (rank==0) printf("Test %d: sum 1 out-of-place\n",count++);
>>>>> in = ( rank==0 ? 1 : 0 );
>>>>> out = 0;
>>>>> ans = 1;
>>>>> MPI_Allreduce(&in,&out,1,MPI_INT,MPI_SUM,MPI_COMM_WORLD);
>>>>> if (out==ans) printf("%4d: in=%d out=%d ans=%d PASSED
>>>>> \n",rank,in,out,ans);
>>>>> else printf("%4d: in=%d out=%d ans=%d FAILED
>>>>> \n",rank,in,out,ans);
>>>>> fflush(stdout);
>>>>>
>>>>> if (rank==0) printf("Test %d: sum 1 in-place\n",count++);
>>>>> in = ( rank==0 ? 1 : 0 );
>>>>> out = in;
>>>>> ans = 1;
>>>>> MPI_Allreduce(MPI_IN_PLACE,&out,1,MPI_INT,MPI_SUM,MPI_COMM_WORLD);
>>>>> if (out==ans) printf("%4d: in=%d out=%d ans=%d PASSED
>>>>> \n",rank,in,out,ans);
>>>>> else printf("%4d: in=%d out=%d ans=%d FAILED
>>>>> \n",rank,in,out,ans);
>>>>> fflush(stdout);
>>>>>
>>>>> if (rank==0) printf("Test %d: sum 2 out-of-place\n",count++);
>>>>> in = 1;
>>>>> out = 0;
>>>>> ans = size;
>>>>> MPI_Allreduce(&in,&out,1,MPI_INT,MPI_SUM,MPI_COMM_WORLD);
>>>>> if (out==ans) printf("%4d: in=%d out=%d ans=%d PASSED
>>>>> \n",rank,in,out,ans);
>>>>> else printf("%4d: in=%d out=%d ans=%d FAILED
>>>>> \n",rank,in,out,ans);
>>>>> fflush(stdout);
>>>>>
>>>>> if (rank==0) printf("Test %d: sum 2 in-place\n",count++);
>>>>> in = 1;
>>>>> out = in;
>>>>> ans = size;
>>>>> MPI_Allreduce(MPI_IN_PLACE,&out,1,MPI_INT,MPI_SUM,MPI_COMM_WORLD);
>>>>> if (out==ans) printf("%4d: in=%d out=%d ans=%d PASSED
>>>>> \n",rank,in,out,ans);
>>>>> else printf("%4d: in=%d out=%d ans=%d FAILED
>>>>> \n",rank,in,out,ans);
>>>>> fflush(stdout);
>>>>>
>>>>> /* TESTING PROD OPERATOR */
>>>>>
>>>>> if (rank==0) printf("Test %d: prod 1 out-of-place\n",count++);
>>>>> in = ( rank==0 ? 1 : 0 );
>>>>> out = 0;
>>>>> ans = 0;
>>>>> MPI_Allreduce(&in,&out,1,MPI_INT,MPI_PROD,MPI_COMM_WORLD);
>>>>> if (out==ans) printf("%4d: in=%d out=%d ans=%d PASSED
>>>>> \n",rank,in,out,ans);
>>>>> else printf("%4d: in=%d out=%d ans=%d FAILED
>>>>> \n",rank,in,out,ans);
>>>>> fflush(stdout);
>>>>>
>>>>> if (rank==0) printf("Test %d: prod 1 in-place\n",count++);
>>>>> in = ( rank==0 ? 1 : 0 );
>>>>> out = in;
>>>>> ans = 0;
>>>>> MPI_Allreduce(MPI_IN_PLACE,&out,1,MPI_INT,MPI_PROD,MPI_COMM_WORLD);
>>>>> if (out==ans) printf("%4d: in=%d out=%d ans=%d PASSED
>>>>> \n",rank,in,out,ans);
>>>>> else printf("%4d: in=%d out=%d ans=%d FAILED
>>>>> \n",rank,in,out,ans);;
>>>>> fflush(stdout);
>>>>>
>>>>> if (rank==0) printf("Test %d: prod 2 out-of-place\n",count++);
>>>>> in = 1;
>>>>> out = 0;
>>>>> ans = 1;
>>>>> MPI_Allreduce(&in,&out,1,MPI_INT,MPI_PROD,MPI_COMM_WORLD);
>>>>> if (out==ans) printf("%4d: in=%d out=%d ans=%d PASSED
>>>>> \n",rank,in,out,ans);
>>>>> else printf("%4d: in=%d out=%d ans=%d FAILED
>>>>> \n",rank,in,out,ans);
>>>>> fflush(stdout);
>>>>>
>>>>> if (rank==0) printf("Test %d: prod 2 in-place\n",count++);
>>>>> in = 1;
>>>>> out = in;
>>>>> ans = 1;
>>>>> MPI_Allreduce(MPI_IN_PLACE,&out,1,MPI_INT,MPI_PROD,MPI_COMM_WORLD);
>>>>> if (out==ans) printf("%4d: in=%d out=%d ans=%d PASSED
>>>>> \n",rank,in,out,ans);
>>>>> else printf("%4d: in=%d out=%d ans=%d FAILED
>>>>> \n",rank,in,out,ans);
>>>>> fflush(stdout);
>>>>>
>>>>> /* END OF TESTS */
>>>>>
>>>>> MPI_Finalize();
>>>>>
>>>>> return 0;
>>>>> }
>>>>> ==========================================================
>>>>>
>>>>
>>>> --
>>>> Pavan Balaji
>>>> http://www.mcs.anl.gov/~balaji
>>>>
>>>
>>>
>>>
>>> --
>>> Jeff Hammond
>>> Argonne Leadership Computing Facility
>>> jhammond at alcf.anl.gov / (630) 252-5381
>>> http://www.linkedin.com/in/jeffhammond
>>>
>>
>>
>>
>> --
>> Jeff Hammond
>> Argonne Leadership Computing Facility
>> jhammond at alcf.anl.gov / (630) 252-5381
>> http://www.linkedin.com/in/jeffhammond
>>
>
>
>
--
Pavan Balaji
http://www.mcs.anl.gov/~balaji
More information about the mpich-discuss
mailing list